|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 34 (December 1), 2006: pp. 5373-5380 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.05.9584 Quantification of Regulatory T Cells Enables the Identification of High-Risk Breast Cancer Patients and Those at Risk of Late Relapse
From the Nuffield Department of Clinical Laboratory Sciences, John Radcliffe Hospital; Cancer Research UK Molecular Oncology Laboratory, Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom; and Monoclonal Antibodies Unit, Biotechnology Program, Centro Nacional de Investigaciones Oncológicas, Madrid, Spain Address reprint requests to Gaynor J. Bates, PhD, Nuffield Department of Clinical Laboratory Sciences, John Radcliffe Hospital, University of Oxford, Oxford, OX3 9DU, United Kingdom; e-mail: gaynor.bates{at}ndcls.ox.ac.uk
PURPOSE: To assess the clinical significance of tumor-infiltrating FOXP3-positive regulatory T cells (TR) in breast cancer patients with long-term follow-up.
PATIENTS AND METHODS: FOXP3-positive TR were detected by immunohistochemistry with our new, extensively characterized FOXP3 monoclonal antibody, 236A/E7. Numbers of FOXP3-positive lymphocytes in tissue microarray cores from pure ductal carcinoma in situ (DCIS; n = 62), invasive breast cancer (n = 237) or from comparable areas of normal terminal duct lobular breast tissue (n = 10) were determined. A median cutoff of
RESULTS: TR numbers were significantly higher in in situ and invasive breast carcinomas than in normal breast; invasive tumors have significantly higher numbers than DCIS (P = .001). High numbers of FOXP3-positive TR identified patients with DCIS at increased risk of relapse (P = .04) and patients with invasive tumors with both shorter relapse-free (P = .004) and overall survival (P = .007). High TR numbers were present in high-grade tumors (P CONCLUSION: These findings indicate that quantification of FOXP3-positive TR in breast tumors is valuable for assessing disease prognosis and progression, and that TR are an important therapeutic target for breast cancer. FOXP3-positive TR represent a novel marker for identifying late-relapse patients who may benefit from aromatase therapy after standard tamoxifen treatment.
CD4+CD25+ regulatory T-cells (TR) are important in the control of immune responses, through their ability to suppress T-cell proliferation and cytokine production.1 They are implicated in a broad spectrum of medical conditions such as autoimmune disease, graft-versus-host disease, allograft rejection, sterilizing immunity to infectious organisms, multiple sclerosis, diabetes, inflammatory bowel disease, and allergy.2 In addition TR have an important role in cancer, enabling tumors to elude the host antitumor immune response. Higher numbers of TR in the peripheral blood of patients with breast,3 liver,4 gastric, and esophageal cancer,5 compared with healthy controls, have been reported. Furthermore, increased numbers of tumor-infiltrating TR have been demonstrated in ovarian,6 hepatocellular,4 gastric, and esophageal cancer.5 Higher TR numbers have also been reported in chronic lymphocytic leukemia patients7; unexpectedly, in Hodgkin's lymphoma, high numbers of accompanying TR have been correlated with an improved outcome.8 Although FOXP3 represents the most specific TR marker, until recently, lack of suitable antibodies has prevented the in vivo labeling of TR in archival formalin-fixed paraffin-embedded tissues. Using specific anti-FOXP3 monoclonal antibodies, we have demonstrated previously that only approximately half the CD4+CD25+ population expressed FOXP3, a minority of FOXP3-positive cells lack CD25 expression, and a small number were CD8+.9 Thus FOXP3-positive TR cells represent a more distinct T-cell population than described in previously published studies relying on a CD4+CD25+ profile alone. We have used immunohistochemistry to assess the clinical significance of the numbers of FOXP3-positive TR in breast tumors. Our prespecified hypothesis was that numbers of FOXP3-positive TR would be clinically relevant in breast cancer patients, with high numbers correlating with a poorer prognosis. We also examined whether there was a relationship between numbers of TR and clinicopathologic data including age, nodal status, tumor size, grade, and human epidermal growth factor receptor-2 (HER2) and estrogen receptor (ER) status. Our study is the first to demonstrate significant relationships between numbers of tumor-infiltrating TR and both clinicopathologic data and survival in this malignancy. Our data are particularly relevant for breast cancer therapy in light of a recent study showing that TR depletion can result in the eradication of a large, well vascularized, highly progressive tumor in mice.10
Patients and Tissue Microarrays The microarrayed tissues from breast carcinomas used in this retrospective study represent a consecutive series of patients from the referral population of a regionally based cancer service and were collected from patients undergoing surgery at The John Radcliffe Hospital (Oxford, United Kingdom). Normal breast tissue was obtained from women undergoing breast reduction surgery (n = 10).
For the previously described11 invasive series of 283 patients, tumors were treated by mastectomy (n = 70) or lumpectomy (n = 213) and axillary node sampling with node status confirmed histologically. Of the 237 tumors on the invasive array, for which FOXP3 staining was assessable, histology was available for 222 patients. Within this group, there were 191 ductal tumors, 15 lobular tumors, and 16 others. Grading was carried out according to the modified Bloom and Richardson method.12 Patients were not matched for stage of disease or age. The only criteria for the selection of patients for the tissue microarray were that there was adequate routinely fixed biopsy material, no history of a previous malignancy, and no distant metastases at presentation. For all ER-positive patients, tamoxifen was prescribed as adjuvant treatment regardless of age or any other prognostic factors. In patients younger than 50 years, adjuvant cyclophosphamide, methotrexate, and fluorouracil was administered if tumors were node positive or ER negative, and/or
For the ductal carcinoma in situ (DCIS) tissue microarray series (n = 120), patients underwent biopsies from 1986 to 1998. Of these, 62 had assessable FOXP3 staining and clinical follow-up. The end of the follow-up period was March 2005 and the median follow-up time was 5.4 years (range, 0.85 to 14.2 years). There were 15 relapses in the DCIS series, which included five ipsilateral DCIS, five ipsilateral invasive, four contralateral invasive tumors, and one metastasis. Twelve patients had a mastectomy and eight patients received radiation. Local ethical approval was obtained for these studies (reference C02.216).
Immunohistochemical Labeling of FOXP3-Positive TR We have confirmed that 236A/E7 specifically recognizes FOXP3 and not other FOXP proteins, and our panel of FOXP3 antibodies was compared by immunohistochemistry on a range of tissues before 236A/E7 was selected for this application. We have published previously the extensive characterization of this antibody,9 and found after comparison with the available commercial reagents that 236A/E7 was the best currently available to study FOXP3 expression by immunohistochemistry.
Quantification of TR
ER and HER2 Status
Statistical Analyses We have also incorporated the newly proposed Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) criteria for tumor marker publications into this study.13 As recommended under the REMARK criteria, we have included a diagram to describe the statistical analysis of patient samples in this study (Fig 1).
In normal breast tissue there were few FOXP3-positive lymphocytes and half of the samples (five of 10) contained no FOXP3-positive TR (median, 0.5; range, 0 to 4). In contrast, although there was a high degree of variability in TR numbers in invasive tumors (median, 15; range, 0 to 579), there were significantly more TR than in normal breast tissues (median, 0.5; range, 0 to 4; Fig 2). TR were localized in the vicinity of the tumor cells, and were also present in lymphoid-enriched areas of the tissue microarray cores (Fig 2D to 2F).
Within the invasive tumor samples, there was a significant correlation between FOXP3-positive TR (  15) and positive lymph node status (P = .01) and higher tumor grade (P  .001). Patients with tumors containing 15 FOXP3-positive TR also had a significantly shorter RFS (P = .004) and OS (P = .007), with hazard ratios of 1.85 (95% CI, 1.21 to 2.84; P = .005) and 1.98 (95% CI, 1.19 to 3.30; P = .008), respectively (Figs 3A and 3B; Table 1). Furthermore, there was a significant negative correlation between TR numbers and ER expression (P = .001). Given that ER-negative tumors generally have a worse prognosis than ER-positive tumors, we determined whether the correlation with TR numbers and survival was purely a reflection of tumor ER status. Within the ER-positive group (Figs 3C and D), patients with tumors containing 15 TR had a significantly shorter RFS (P = .005) and OS (P = .005), with hazard ratios of 2.20 (95% CI, 1.26 to 3.85; P = .006) and 2.57 (95% CI, 1.31 to 5.60; P = .006), respectively. In contrast, TR numbers had no significant impact on survival within the ER-negative group (Figs 3E and 3F). Interestingly, there was no significant difference between the survival of ER-positive versus ER-negative patients with 15 TR (P = .52 and P = .62 for RFS and OS, respectively). Quantification of FOXP3-positive TR identified a group at high risk of relapse within a so-called good prognostic group of ER-positive patients. Indeed, these patients have a prognosis as poor as those who lack ER expression. In multivariate analyses, high TR numbers conferred a significantly shorter RFS (P = .04; Table 2). Multivariate analyses in ER-positive patients demonstrated that greater TR numbers independently conferred a significantly higher hazard ratio than that of tumor grade and nodal status for RFS and OS, respectively (Table 3).
We also investigated the prognostic significance of numbers of TR separately within the ER-positive subgroup of ductal tumors. In the ER-positive tumors (n = 117; relapses, n = 37; deaths, n = 26) multivariate analysis showed that higher numbers of TR independently conferred a significantly reduced RFS and OS (P = .03), and high numbers of TR conferred a significantly higher hazard ratio (2.15; 95% CI, 1.10 to 4.17; P = .03) than that of tumor grade (1.88; 95% CI, 1.13 to 3.13; P = .02) for RFS (Table 4). However, the numbers of relapses and deaths in this subgroup of patients with ER-positive ductal tumors were relatively small. If ER-positive and ER-negative ductal tumors were evaluated together (n = 191; relapses, n = 70; deaths, n = 51), high numbers of TR showed borderline significance for OS (P = .07) and were significant for RFS (P = .03). However, this was no longer significant in a multivariate model after controlling for the effects of nodal status, tumor size, grade, and ER status (P = .13 and 0.32 for OS and RFS, respectively).
There was a significant relationship between HER2 expression and TR numbers (P = .02; Table 1) although because few of the patients with invasive disease in this series were HER2 positive, this marker was not significantly associated with survival in multivariate analyses (data not shown). We also investigated if there was any relationship between numbers of TR and whether patients received adjuvant chemotherapy. In univariate analysis, a greater number of patients with high numbers of TR received adjuvant chemotherapy compared with those with low numbers (P = .01; Table 5). However, in multivariate analysis this was no longer significant. There was no significant relationship between TR numbers and whether patients received adjuvant radiotherapy, hormone therapy, or radical surgery.
Patient groups treated with either chemotherapy or hormonal therapy were also stratified for survival based on TR numbers to determine whether these might predict relative response or resistance to treatment. Only 21 patients were treated with chemotherapy without hormone treatment. In this group there were eight deaths and 12 relapses, and TR numbers did not correlate with either OS (P = .9) or RFS (P = .29). There were 131 patients treated with tamoxifen, without chemotherapy; 28 deaths and 42 relapses occurred in this group. In this group, high TR numbers were significantly associated with both adverse OS (P = .01) and RFS (P = .01) in univariate analyses. However, this was not significant in multivariate analysis including nodal status and grade (OS, P = .21; RFS, P = .26). We also investigated if the prognostic value of TR varied during the follow-up period. Interestingly, we found that TR numbers only marginally predicted outcome at 3 years after diagnosis (RFS, P = .06; OS, P = .05). In patients who were disease free and who had not experienced local or systemic relapsed by 5 years, FOXP3-positive TR numbers enabled the identification of those patients at risk of late relapse after 5 years (P = .05; Fig 4). This is important given that, within our series, no conventional clinicopathological factors predicted relapse after 5 years (size, nodes, grade, or ER), and therefore FOXP3-positive TR are a novel clinical biomarker for the late relapse of this group of breast cancer patients. Furthermore, we show that there is an increasing separation per year in the annual hazard ratios between those patients who have high numbers of TR compared with those with low numbers, throughout the entire follow-up period (Fig 4).
Finally, to investigate whether TR numbers were also prognostic in noninvasive breast cancer, we analyzed the numbers of FOXP3-positive TR in 62 DCIS patients. Interestingly, the median TR number differed significantly between normal, DCIS, and invasive breast tumors (normal, 0.5; DCIS, four; invasive, 15; P = .001; range, zero to four, zero to 161, and zero to 579, respectively), suggesting that TR accumulation represents a marker of breast cancer progression. Higher numbers of TR in DCIS patients (Fig 5) also indicated a worse RFS (P = .04), conferring a hazard ratio of 2.81 (95% CI, 0.99 to 7.99; P = .05). Patients with higher grade tumors contained higher TR numbers; however, this was not significant and the correlation between higher TR numbers and RFS was no longer significant after controlling for the effect of grade (P = .25). Given that only two deaths occurred in the DCIS series, it was not possible to correlate numbers of TR with OS.
Although previous studies reported an increased pool of CD4+CD25+ TR in the peripheral blood of a small number of breast cancer patients, before our current study the clinical importance of tumor-infiltrating FOXP3-positive TR in diagnostic biopsies from this malignancy was unknown. We observed significant correlations with the more aggressive tumor phenotype of ER negativity, high tumor grade, and positive nodal status. The observation that elevated TR numbers confer a significantly shorter OS and RFS in invasive breast tumors, described here, is consistent with the report of increased TR numbers in high-risk patients with ovarian carcinoma.6 The data are also consistent with the widely accepted hypothesis that the recruitment of tumor-infiltrating TR may enable malignant cells to evade the host immune response. Our novel identification of a high-risk subgroup of patients with ER-positive tumors containing high numbers of TR, who have as poor a prognosis as patients with tumors that lack ER expression, is particularly important because these patients currently would be assessed as having a good prognosis. In contrast, given that ER-negative patients have such a poor prognosis compared with those with ER-positive tumors, it is not surprising that numbers of TR do not influence survival in this particularly high-risk patient group. Although standard prognostic and pathologic factors predict relapse in the first 5 years after therapy (size, grade, nodal status, and necrosis), many longer-term follow-up studies suggest they are no longer significant when applied to those who survive 5 years.14-16 Furthermore, molecular markers such as ER, progesterone receptor, epidermal growth-factor receptor 1, and HER2 have only been related to early rather than late relapse.17,18 Thus there is a clear need to identify novel markers for late relapse in breast cancer patients, the mechanisms of which are poorly understood. We hypothesized that immunologic factors may be relevant, particularly relating to longer-term tumor dormancy. We therefore evaluated the prognostic significance of FOXP3-positive TR in patients who survived to 5 years without any local or systemic relapse. We present the novel finding that high FOXP3-positive TR numbers represent an important marker for the identification of breast cancer patients at risk of late relapse. Furthermore, it has been demonstrated recently that treatment with letrozole, an aromatase inhibitor, after the completion of 5 years standard tamoxifen treatment significantly improves disease-free survival.19 Given that TR numbers and functionality have been shown to be increased by estrogen,20,21 it is possible that this aromatase inhibitor, through suppression of estrogen production, may be able to reduce the number of tumor-infiltrating TR and result in the improved patient disease-free survival, after completion of tamoxifen therapy. Additional studies are currently underway to test this hypothesis. TR numbers potentially represent a novel marker for identifying patients who would benefit from the administration of aromatase therapy after 5 years treatment with tamoxifen. In conclusion, the number of tumor-associated TR is a significant parameter for disease prognosis in both invasive and noninvasive breast tumors that can be assessed in routinely fixed tissues by immunohistochemistry to detect FOXP3-positive T cells. Furthermore, tumor vaccination approaches, in combination with targeting TR cells, are just entering clinical trials22 and our data strongly suggest that such therapy would be beneficial for a significant proportion of breast cancer patients. Given the broad spectrum of diseases in which TR have been implicated, the tools that are now in place to enable their quantification should have a significantly wider multidisciplinary impact in clinical research.
The authors indicated no potential conflicts of interest.
Supported by Breast Cancer Campaign, Cancer Research UK and Leukaemia Research Fund. Presented in part at the 21st Genes and Cancer Meeting, December 13-15, 2004, University of Warwick, United Kingdom; and at the 5th European Breast Cancer Conference, March 21-25, 2006, Nice, France. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Thornton AM, Shevach EM: CD4+CD25+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J Exp Med 188:287-296, 1998 2. Thompson C, Powrie F: Regulatory T cells. Curr Opin Pharmacol 4:408-414, 2004[CrossRef][Medline] 3. Liyanage UK, Moore TT, Joo HG, et al: Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocarcinoma. J Immunol 169:2756-2761, 2002 4. Ormandy LA, Hillemann T, Wedemeyer H, et al: Increased populations of regulatory T cells in peripheral blood of patients with hepatocellular carcinoma. Cancer Res 65:2457-2464, 2005 5. Ichihara F, Kono K, Takahashi A, et al: Increased populations of regulatory T cells in peripheral blood and tumor-infiltrating lymphocytes in patients with gastric and esophageal cancers. Clin Cancer Res 9:4404-4408, 2003 6. Curiel TJ, Coukos G, Zou L, et al: Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 10:942-949, 2004[CrossRef][Medline] 7. Beyer M, Kochanek M, Darabi K, et al: Reduced frequencies and suppressive function of CD4+ CD25 high regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood 106:2018-2025, 2005 8. Alvaro T, Lejeune M, Salvado MT, et al: Outcome in Hodgkin's lymphoma can be predicted from the presence of accompanying cytotoxic and regulatory T cells. Clin Cancer Res 11:1467-1473, 2005 9. Roncador G, Brown PJ, Maestre L, et al: Analysis of FOXP3 protein expression in human CD4(+)CD25(+) regulatory T cells at the single-cell level. Eur J Immunol 35:1681-1691, 2005[CrossRef][Medline] 10. Yu P, Lee Y, Liu W, et al: Intratumor depletion of CD4+ cells unmasks tumor immunogenicity leading to the rejection of late-stage tumors. J Exp Med 201:779-791, 2005 11. Fox SB, Brown P, Han C, et al: Expression of the forkhead transcription factor FOXP1 is associated with estrogen receptor alpha and improved survival in primary human breast carcinomas. Clin Cancer Res 10:3521-3527, 2004 12. Elston C: Grading of Invasive Carcinoma of the Breast. Edinburgh, United Kingdom, Churchill Livingstone, 1987 13. McShane LM, Altman DG, Sauerbrei W, et al: Reporting recommendations for tumor marker prognostic studies. J Clin Oncol 23:9067-9072, 2005 14. Langlands AO, Pocock SJ, Kerr GR, et al: Long-term survival of patients with breast cancer: A study of the curability of the disease. BMJ 2:1247-1251, 1979 15. Toikkanen SP, Kujari HP, Joensuu H: Factors predicting late mortality from breast cancer. Eur J Cancer 27:586-591, 1991[Medline] 16. Gilchrist KW, Gray R, Fowble B, et al: Tumor necrosis is a prognostic predictor for early recurrence and death in lymph node-positive breast cancer: A 10-year follow-up study of 728 Eastern Cooperative Oncology Group patients. J Clin Oncol 11:1929-1935, 1993 17. Collett K, Skjaerven R, Maehle BO: The prognostic contribution of estrogen and progesterone receptor status to a modified version of the Nottingham Prognostic Index. Breast Cancer Res Treat 48:1-9, 1998[CrossRef][Medline] 18. Tovey S, Dunne B, Witton CJ, et al: Can molecular markers predict when to implement treatment with aromatase inhibitors in invasive breast cancer? Clin Cancer Res 11:4835-4842, 2005 19. Goss PE, Ingle JN, Martino S, et al: A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 349:1793-1802, 2003 20. Polanczyk MJ, Carson BD, Subramanian S, et al: Cutting edge: Estrogen drives expansion of the CD4+CD25+ regulatory T cell compartment. J Immunol 173:2227-2230, 2004 21. Prieto GA, Rosenstein Y: Oestradiol potentiates the suppressive function of human CD4 CD25 regulatory T cells by promoting their proliferation. Immunology 118:58-65, 2006[CrossRef][Medline] 22. Emens LA, Reilly RT, Jaffee EM: Breast cancer vaccines: Maximizing cancer treatment by tapping into host immunity. Endocr Relat Cancer 12:1-17, 2005 Submitted February 1, 2006; accepted September 25, 2006.
Related Correspondence
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test. Survival was measured from the date of diagnosis to the time of death/relapse or the time the patient was last seen. The log-rank test was used to perform univariate analyses and the survival curves were estimated by the Kaplan-Meier method. Prognostic factors for survival were evaluated in multivariate analyses by Cox proportional hazards regression. The statistical tests are detailed in the relevant figure legends. 
